|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 36 (December 20), 2005: pp. 9429-9432 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.04.2390
In Reply:Department of Internal Medicine I, University of Bonn, Germany
The Royal Free Hospital, London, United Kingdom
Department of Internal Medicine I, University of Bonn, Germany
Institute for Hygiene and Public Health, University of Bonn, Germany
Department of Internal Medicine I, University of Bonn, Germany
Department of Interdisciplinary Oncology, H. Lee Moffitt Cancer Center & Research Institute, University of South Florida, Tampa, FL
Department of Internal Medicine I, University of Bonn, Germany The response of Gøtzsche and Johansen to our meta-analysis,1 which criticized their use of the same methodology in the same area, is interesting. Our comments related solely to methodological features of their meta-analysis on antifungal prophylaxis.2,3 Naturally, we are sorry if our criticisms upset them personally, but feel it is necessary to repeat and enlarge on them now in view of their letter. Empirical antifungal treatment is a completely different concept and clinical practice from antifungal prophylaxis, the former attempting to treat something which cannot be proved to be present at the start of treatment, and the latter attempting to prevent the same problem occurring when that is thought to be a high risk. This is widely accepted and understood (including by investigating clinicians, ethics committees, and regulatory authorities like United States Food and Drug Administration and European Medicines Agency) in the design and conduct of clinical trials as governed by good clinical practice. In the management of systemic fungal infection and in the analysis of results of clinical trials in these infections this distinction is critical. In addition, in view of the many flaws in the design and conduct of therapeutic trials in systemic fungal infection it is even more necessary to exclude them from any analysis of the outcomes of prophylaxis trials. Gøtzsche and Johansen lump prophylaxis and therapeutic trials together in their meta-analysis of the efficacy of antifungal drugs. The lack of differentiation between these two indications is one of our core criticisms of Gøtzsche and Johansen’s approach. In their original publication2 and in the latest version of their review, (published in issue 4, 2004, of the Cochrane Library3), three trials of empirical amphotericin B therapy4-6 are included with five prophylaxis studies. These studies were pooled in their meta-analysis for the outcome of death (Fig 1) and for the outcome invasive infections (Fig 2). Similarly but more worryingly a trial of empiric therapy of fluconazole7 is called a trial of prophylaxis. These basic errors invalidate their conclusions not only for each of these drugs individually and for the summation of all trials of all drugs but also their conclusions for the strategies of antifungal prophylaxis and therapy separately. Gøtzsche and Johansen fail in their meta-analysis to take into account key characteristics of the drugs used in the trials they analyzed. In the trials of amphotericin B they include one using oral amphotericin B as active prophylaxis8 in a dose which might not even achieve decontamination of the gut lumen and could not achieve any relevant systemic bioavailability. This could be described as a placebo versus placebo study and including it in the analysis of amphotericin B trials invalidates further their conclusions about the efficacy of this drug and all the drugs together in their meta-analysis. Even in the trials of intravenous amphotericin B that they included, the doses used could be described as inadequate. Adequate bioavailability of the antifungal drug is a key issue as our meta-analysis reveals. Gøtzsche and Johansen claim that we have not included trials in our meta-analysis "where amphotericin B was given intravenously" without quoting any references. Presumably they mean comparisons of prophylactic intravenous amphotericin B with prophylactic itraconazole although they do not make that clear. We did not include such studies with intravenous amphotericin B as a comparator in our meta-analysis because we could not find any by repeated searches of electronic databases, volumes of abstracts, and inquiries to the manufacturer of the drug. It is possible, but unlikely, that we have missed one or more. We feel strongly that Gøtzsche and Johansen are obliged to reveal the references for these trials if they have them. To our knowledge the only studies that compare these drugs have been done in empirical antifungal therapy and not in antifungal prophylaxis.9,10 We have analyzed such trials separately as is appropriate and presented the results.11 There is insufficient evidence to justify the statement that either oral (nonabsorbable, therefore nonsystemic) or systemic amphotericin B provides effective prophylaxis against systemic fungal infection in hematological malignancy and further prophylaxis trials of this drug given systemically in its various formulations would be of interest. Gøtzsche and Johansen say correctly that we do not mention the efficacy of voriconazole in our paper. That is because there are as yet no reports of randomized controlled trials of its use in prophylaxis of systemic fungal infection. One is in progress and others are being discussed with the manufacturer. The results will be of great interest. Gøtzsche and Johansen fail to take into account the important differences in the spectra of antifungal activity of these drugs and this further invalidates their meta-analysis. Even among individual classes where the unwary might expect some class action justifiying combined meta-analysis, such as the azoles, there are critical differences. For example fluconazole has never been shown to be clinically active against Aspergillus spp. Therefore it cannot be lumped together in a meta-analysis of antifungal efficacy with itraconazole. To avoid these mishaps in future it could be more productive if content-specific and method-specific experts cooperated as in our meta-analysis than if method-specific experts worked alone. This could perhaps be of more benefit to our patients than abusive criticism of one group by another. Gøtzsche and Johansen claim that we have not analyzed a large unpublished trial that was mentioned by one of us in the literature.12 We can understand why they may have been mislead by this statement, but such misunderstanding could have been avoided through the cooperation described above. This trial, however, as confirmed by both investigators and the pharmaceutical manufacturer, never recruited to its original design. It was modified to include an active comparison arm (fluconazole) and to switch from capsules to solution of itraconazole. The results were published and included in our meta-analysis.13 There were good reasons for the changes in this trial. At that time many randomized controlled trials of fluconazole were showing such a significant benefit in preventing systemic candidiasis14 that a placebo controlled trial might not be seen as ethical. The pharmacokinetic studies conducted by one of us15,16 were showing that the bioavailability of itraconazole solution was superior to the capsule forms. Our meta-analysis confirms in retrospect that this was a logical decision. At that time an unpublished trial of prophylaxis with itraconazole capsules (mentioned in17) had already been abandoned. This trial was stopped early because plasma concentrations achieved with itraconazole capsules were low or zero in an initial small number of patients. The results of this unfinished trial are irrelevant to—and would not have changed—the results of our meta-analysis. Overall, our efforts to provide all clinical trials for the meta-analysis resulted in an intensive interaction with the manufacturer of itraconazole, which lead to the inclusion of unpublished data.19 An ethical, science-based and, at times, critical co-operation with the pharmaceutical industry is in our eyes more appropriate to achieve better patient care than a general condemnation. We agree that tolerability is an important aspect of antifungal prophylaxis. Our meta-analysis confirmed that hypokalaemia with all formulations of the drug and nausea with the oral solution were the only clinically important adverse effects. Congestive heart failure as a result of itraconazole therapy was not reported in any of the trials analyzed. Gøtzsche and Johansen also remark that azoles may be associated with increased deaths due to bacterial infections. This has been discussed controversially in the literature.18,20,21 In our meta-analysis two studies reported on the incidence of bacterial infections (one with a statistically significant outcome and one not) and a quantitative synthesis of their results did not demonstrate a significantly higher rate of bacterial infections (odds ratio = 1.23, 95%CI, 0.82 to 1.84).22,23 Further study of this question is certainly needed. Gøtzsche and Johansen claim that only a reduction of the overall mortality could justify the use of antifungal prophylaxis. We disagree with this contention as the occurrence of an invasive fungal infection—the primary outcome parameter of our analysis and of the included trials—is a disastrous, life-threatening, and costly complication for the affected patient. The widely reported death rate from established aspergillosis in patients being treated for hematological malignancy justifies fully the use of effective antifungal prophylaxis and the conduct of further trials to improve on that efficacy.24 Prophylaxis with itraconazole solution reduced the incidence of these infections by 48%, in our systematic review. Patients and hospitals clearly benefit from the reduction of this dangerous and costly infection. None of the trials included in our meta-analysis were designed or powered to detect a difference in overall survival. One study reported survival benefits with fluconazole prophylaxis in stem-cell transplant recipients but only after prolonged prophylaxis (75 days) and observation times (years).25 One other study with itraconazole has shown that prolonged observation demonstrates a higher efficacy of antifungal prophylaxis (we have, however, excluded this data as it occurred outside of the originally defined study period).13
Several times in their letter, Gøtzsche and Johansen draw the results of our meta-analysis into doubt because it was supported in part by the manufacturer of itraconazole (in 2000-2003, the University of Bonn received research grants in total of We are well aware of biases that may be associated with industry sponsorship.26 This awareness is exactly the reason that we have approached this meta-analysis (or any other of our meta-analyses) with extreme caution and with maximum attempts to minimize any bias that may be introduced in the analysis. We believe that we minimized these biases and consequently preserved the validity of our study. Therefore, Gøtzsche and Johansen should not make such an accusation without proof, as it appears that they set themselves up as our judge and jury and consider all clinicians guilty without trial if they cooperate with the pharmaceutical industry. The reality is much more complex than such a simple and overzealous view of a malign influence of the pharmaceutical industry. Instead, we should try to dissect the mechanism of industry-sponsored research and undertake measures to minimize these biases. This includes using the approach exercised in our meta-analysis, which among other things included coauthors with no association with the manufacturer of itraconazole whatsoever. We would certainly agree with many critics that the relationship between the industry, the investigators, and the regulators can be and often is dysfunctional. However, this relationship can and must be improved. For financial and other reasons clinical, patient-oriented research may not be sustainable without close collaboration between all three of these groups, and unjustified abuse from nonpracticing critics could simply lead to their exclusion from that collaboration. The history of science demonstrates that bias has many other sources than financial support. Our minds and our ethical practice of medicine and clinical trials have not been and cannot be bought by the pharmaceutical industry. We performed our meta-analysis not at their request, but because we thought that this methodology would be the most rigorous test of results of trials of the efficacy of this drug. The control over the analysis, the content of the paper, and the decision to submit the manuscript were always in our hands and in no way influenced by the pharmaceutical industry. Our meta-analysis can be reanalyzed by any reader who is willing to invest sufficient time and we will support any reasonable inquiries on our data. We uphold our methodological criticism of Gøtzsche and Johansen’s meta-analysis and the results of our meta-analysis. We recommend antifungal prophylaxis with sufficient doses of itraconazole for patients with myelosuppressive chemotherapy or after allogeneic stem cell transplantation. Nowhere in our paper do we conclude or recommend without reservation "that immunosuppressed cancer patients should receive an antifungal prophylaxis with itraconazole" as Gøtzsche and Johansen try to summarize our conclusions. This gross distortion illustrates another failure of Gøtzsche and Johansen to understand this complex field and it will help readers to draw their own conclusions about the relative merits of our work. We apologize for the length of this detailed response. Authors’ Disclosures of Potential Conflicts of Interest
The author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed discription of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Dollar Amount Codes (A) < $10,000 (B) $10,000–99,000 (C)
REFERENCES
1. Glasmacher A, Prentice AG, Gorschluter M, et al: Itraconazole prevents invasive fungal infections in neutropenic patients treated for haematological malignancies: Evidence from a meta-analysis of 3,597 patients. J Clin Oncol 21:4615-4626, 2003 2. Gotzsche PC, Johansen HK: Meta-analysis of prophylactic or empirical antifungal treatment versus placebo or no treatment in patients with cancer complicated by neutropenia. BMJ 314:1238-1244, 1997 3. Gotzsche PC, Johansen HK: Routine versus selective antifungal administration for control of fungal infections in patients with cancer. Cochrane Database Syst Rev CD000026, 2002 4. Pizzo PA, Robichaud KJ, Gill FA, et al: Empiric antibiotic and antifungal therapy for cancer patients with prolonged fever and granulocytopenia. Am J Med 72:101-111, 1982[CrossRef][Medline] 5. EORTC International Antimicrobial Therapy Cooperative Group: Empiric antifungal therapy in febrile granulocytopenic patients. EORTC International Antimicrobial Therapy Cooperative Group. Am J Med 86:668-672, 1989[CrossRef][Medline] 6. Goldstone AH, O’Driscoll A: Early AmBisome in febrile neutropenia in patients with haematological disorders. Bone Marrow Transplant 14:15-17, 1994[Medline] 7. Fukuda M, Hirashima K, Kurane R, et al: Empiric therapy with fluconazole in granulocytopenic patients with carcinoma or leukemia. Jpn J Antibiot 47:1065-1070, 1994[Medline] 8. Suda T, Omine M, Tsuchiya J, et al: A co-operative study on prophylaxis of fungal infection in patients with hematological diseases: Prophylactic effect of oral administration of amphotericin B (author’s transl). Rinsho Ketsueki 21:195-199, 1980[Medline] 9. Boogaerts M, Winston DJ, Bow EJ, et al: Intravenous and oral itraconazole versus intravenous amphotericin B deoxycholate as empirical antifungal therapy for persistent fever in neutropenic patients with cancer who are receiving broad-spectrum antibacterial therapy. A randomized, controlled trial. Ann Intern Med 135:412-422, 2001 10. Ehninger G, Schuler U, Bammer S, et al: Intravenous itraconazole followed by oral itraconazole versus intravenous amphotericin B as empirical antifungal therapy for febrile neutropenic haematological cancer patients. Blood 100:42b, 2002 (suppl 2; abstr 3631) 11. Glasmacher A, Boogaerts M, Schuler U, et al: Combined analysis of two randomized, controlled trials comparing empirical antimycotic therapy with intravenous itraconazole or amphotericin B in neutropenic patients with persistent fever. Proc Intersc Conf Antimicrob Agents Chemother 413: 2004 (abstr M-1016) 12. Prentice AG, Bradford GR: Prophylaxis of fungal infections with itraconazole during remission-induction therapy. Mycoses 32:96-102, 1989 (suppl 1) 13. Morgenstern GR, Prentice AG, Prentice HG, et al: A randomized controlled trial of itraconazole versus fluconazole for the prevention of fungal infections in patients with haematological malignancies: U.K. Multicentre Antifungal Prophylaxis Study Group. Br J Haematol 105:901-911, 1999[CrossRef][Medline] 14. Kanda Y, Yamamoto R, Chizuka A, et al: Prophylactic action of oral fluconazole against fungal infection in neutropenic patients: A meta-analysis of 16 randomized, controlled trials. Cancer 89:1611-1625, 2000[CrossRef][Medline] 15. Prentice AG, Warnock DW, Johnson SA, et al: Multiple dose pharmacokinetics of an oral solution of itraconazole in patients receiving chemotherapy for acute myeloid leukaemia. J Antimicrob Chemother 36:657-663, 1995 16. Prentice AG, Warnock DW, Johnson SA, et al: Multiple dose pharmacokinetics of an oral solution of itraconazole in autologous bone marrow transplant recipients. J Antimicrob Chemother 34:247-252, 1994 17. Bradford CR, Prentice AG, Warnock DW, et al: Comparison of the multiple dose pharmacokinetics of two formulations of itraconazole during remission induction for acute myeloblastic leukaemia. J Antimicrob Chemother 28:555-560, 1991 18. Viscoli C, Paesmans M, Sanz M, et al: Association between antifungal prophylaxis and rate of documented bacteremia in febrile neutropenic cancer patients. Clin Infect Dis 32:1532-1537, 2001[CrossRef][Medline] 19. Glasmacher A, Cornely O, Ullmann AJ, et al: An open-label randomised trial comparing itraconazole oral solution with fluconazole oral solution for primary prophylaxis of fungal infections in patients with haematological malignancy and profound neutropenia. Journal of Antimicrobial Chemotherapy 2006, in press 20. Viscoli C, Paesmans M: Antifungal prophylaxis and the rate of bacteremia among neutropenic patients. Clin Infect Dis 34:289-291, 2002[Medline] 21. Wenzel RP, Gennings C, Edmond MB: Antifungal antibiotics and breakthrough bacteremias. Clin Infect Dis 32:1538-1539, 2001[CrossRef][Medline] 22. Huijgens PC, Simoons-Smit AM, van Loenen AC, et al: Fluconazole versus itraconazole for the prevention of fungal infections in haemato-oncology. J Clin Pathol 52:376-380, 1999[Abstract] 23. Menichetti F, Del Favero A, Martino P, et al: Itraconazole oral solution as prophylaxis for fungal infections in neutropenic patients with hematologic malignancies: A randomized, placebo-controlled, double-blind, multicenter trial. GIMEMA Infection Program. Gruppo Italiano Malattie Ematologiche dell’ Adulto. Clin Infect Dis 28:250-255, 1999[Medline] 24. Lin SJ, Schranz J, Teutsch SM: Aspergillosis case-fatality rate: Systematic review of the literature. Clin Infect Dis 32:358-366, 2001[CrossRef][Medline] 25. Marr KA, Seidel K, Slavin MA, et al: Prolonged fluconazole prophylaxis is associated with persistent protection against candidiasis-related death in allogeneic marrow transplant recipients: Long-term follow-up of a randomized, placebo-controlled trial. Blood 96:2055-2061, 2000 26. Lexchin J, Bero LA, Djulbegovic B, et al: Pharmaceutical industry sponsorship and research outcome and quality: Systematic review. BMJ 326:1167-1170, 2003
Related Correspondence
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
12.400 for all projects with itraconazole, specific funding for this meta-analysis was restricted to the cost of preparing the figures). The pharmaceutical manufacturer did not influence the decision to publish, the analysis, or the preparation of the manuscript in any other way than providing data at our request. 
$100,000 (N/R) Not Required
